The present disclosure relates generally to a radio frequency (RF) receivers, and more particularly to RF receivers subject to magnetic coupling of signals between circuits.
Radio frequency RF) receivers are used in a wide variety of applications such as television receivers, cellular telephones, pagers, global positioning system (UPS) receivers, cable moderns, cordless phones, satellite radio receivers, and the like. As used herein, a “radio frequency” signal means an electrical signal conveying useful information and having a frequency from about 3 kilohertz (kHz) to hundreds of gigahertz (GHz), regardless of the medium through which such signal is conveyed. Thus an RF signal may be transmitted through air, free space, coaxial cable, fiber optic cable, etc. One common type of RF receiver is the so-called superheterodyne receiver. A superheterodyne receiver mixes the desired data-carrying signal with the output of tunable oscillator to produce an output at a fixed intermediate frequency (IF). The fixed IF signal can then be conveniently filtered and converted down to baseband for further processing. Thus a superheterodyne receiver requires two mixing steps.
Modern integrated circuit technology has allowed many of the circuits used in RF receivers to be combined on-chip and thus to substantially reduce the cost of the RF receiver. However this level of integration creates other problems. For example, signals from one part of the chip may be electrically or magnetically coupled to circuits in another part of the chip. These unwanted signal couplings can distort the desired signal and create artifacts that can be perceived by the viewer or listener. Traditionally, integrated circuit designers have used layout strategies to reduce coupling between circuits, such as physical separation, the addition of ground rings, a reduction in the length of conductors, etc. However these techniques, while still useful, are unable to completely eliminate the deleterious effects of electrically or magnetically coupled energy.
The present disclosure may be better understood, and its numerous features and advantages made apparent to those skilled in the art by referencing the accompanying drawings, in which:
The use of the same reference symbols in different drawings indicates similar or identical items.
TV tuner chip 120 includes an analog receiver 122, a digital processor 124, and a clock synthesizer 126, Analog receiver 122 has an input coupled to antenna 110 through a balanced-unbalanced (balun) transformer, not shown in
Digital processor 124 includes a digital signal processor (DSP) 130 and a microcontroller (MCU) 132. DSP 130 has a signal input for receiving the IF signal, a clock input for receiving a clock signal labeled “fCLK”, and an output for providing signal IF OUTPUT, MCU 132 has an input for receiving the CHANNEL SELECTION input, a clock input for receiving fCLK, a first output for providing the TUNING CONTROL SIGNAL, and a second output for providing a signal labeled “CLOCK CONTROL SIGNAL”.
Clock synthesizer 126 has two terminals connected to crystal 140, a control input for receiving the CLOCK CONTROL SIGNAL from MCU 132, a first output connected to the clock input of analog receiver 122, and a second output for providing signal fCLK.
Generally, TV tuner chip 120 operates to tune a selected television channel and output the channel to digital demodulator 150. Analog receiver 122 tunes the selected channel by mixing it with a local oscillator signal whose frequency is chosen to mix the selected channel to a desired IF. In TV tuner chip 120, the desired IF is a low IF of 4 MHz, but in TV tuner chip 120 the IF may also be chosen as a zero IF. Analog receiver 122 further processes the analog IF signal and converts it to digital form using analog-to-digital converters (ADCs), not shown, fix further processing in the digital domain by DSP 130. In addition to supporting DIV receivers as illustrated, TV tuner chip 120 also supports analog television (ATV) receivers and integrates an ATV demodulator, not shown in
MCU 132 performs frequency planning under the control of firmware. It converts the CHANNEL SELECTION input into an appropriate CLOCK CONTROL SIGNAL to clock synthesizer 126. MCU 132 includes a memory, such as a read-only memory (ROM) or FLASH memory, for storage of the firmware that controls the frequency planning algorithm, as well as a serial input/output port for receiving the CHANNEL SELECTION input.
Because it integrates circuitry operating at different frequencies, TV tuner chip 120 is susceptible to the harmful effects of undesired signal coupling, known as spurs. A spur is any unwanted tonal energy generated by the device itself. A spur can harm the signal in the wanted channel either by acting as a co-channel, or if it corrupts the VCO, it can reciprocally mix with large blockers and hence hurt the wanted channel. However TV tuner chip 120 performs frequency planning for fCLK to take into account the effects of three main spur coupling mechanisms. First, it takes into account the direct coupling of fCLK and its harmonics to other circuits. Digital processor 124 operates in synchronization with fCLK. As is conventional, fCLK is a square wave signal that has significant energy at its nominal frequency as well as certain harmonic frequencies. If these signals are coupled to circuits that tune or process the RF INPUT signal, undesired energy could be coupled into the passband of analog processor 122, distorting the signal.
Second, fCLK can mix with a signal generated in clock synthesizer 126 to create a spur. The spur is harmful when its frequency overlaps with the frequency of the desired channel.
Another spur mechanism is the generation and coupling of sub-harmonics. A sub-harmonic is a signal with energy that is generated at a fraction of a given frequency such as fCLK and at harmonics of this fraction. In TV tuner chip 120, digital processor 124 operates in synchronism with the fCLK signal. Certain events, such as the switching of internal buses, occur at intervals equal to an integer number of fCLK periods. These switching events create energy that appears at frequencies tower than fCLK. What makes this mechanism particularly insidious is that it is difficult to identify the primary sub-harmonic by using computer simulation alone, but may require observation on the chip itself after the design is complete. For example in designing TV tuner chip 120, the inventors estimated from computer simulation that digital processor 124 would generate a primary sub-harmonic at fCLK/8 (and other sub-harmonics at harmonics of fCLK/8). However when observing the manufactured chip, they discovered the primary sub-harmonic to actually occur at fCLK/16.
By using frequency planning, TV tuner chip 120 is able to change its frequency to any suitable frequency in the range of clock synthesizer 126. In order to support frequency planning, TV tuner chip 120 includes a clock synthesizer, clock synthesizer 126, that is capable of changing its operating frequency from a nominal operating frequency to frequencies within an available frequency range in small, discrete steps. In this way MCU 132, under the control of firmware stored in its internal program memory, is able to change fCLK to a frequency that reduces undesired coupling.
The techniques by which MCU 132 performs frequency planning for fCLK will now be described with reference to
Crystal oscillator circuit 210 includes two terminals connected to respective terminals of crystal 140. Crystal oscillator circuit 210 inverts the signal between the terminals of crystal 140 to create oscillation and to provide a clock output signal to analog receiver 122. In the illustrated embodiment, crystal 140 is a low-cost crystal having a standard frequency of 24 megahertz (MHz). Note that analog receiver 122 additionally multiplies the 24 MHz clock signal according to the TUNING CONTROL SIGNAL to form a local oscillator signal to mix the selected channel to the desired IF.
Divider 220 has an input connected to the output of crystal oscillator circuit 210, and an output. In the illustrated embodiment, divider 220 divides the 24 MHz output signal by six to produce a 4 MHz signal.
PLL 230 converts the 4 MHz signal to a multiplied signal. PLL 230 is a digital PLL, that includes a phase detector 232, a loop filter 234, a voltage controlled oscillator (VCO) 236, and a feedback divider 238. Phase detector 232 has a positive input connected to the output of divider 220, a negative input, and an output. Loop filter 234 has an input connected to the output of phase detector 232, and an output. VCO 236 has an input connected to the output of loop filter 234, and an output for providing a multiplied signal. Feedback divider 238 has an input connected to the output of VCO 236, a control input for receiving the CLOCK CONTROL SIGNAL, and an output connected to the negative input of phase detector 232. Feedback divider 238 changes the divide ratio of the VCO output signal according to the CLOCK CONTROL SIGNAL and in conjunction with phase detector 232 allows VCO 235 to provide the multiplied signal in 4 MHz frequency steps.
In the illustrated embodiment, the output of PLL 230 is a signal with a nominal frequency of 1.6 gigahertz (GHz) but which may be varied in increments of 4 MHz. Divider 240 is a fixed divider that divides the input thereof by eight. As illustrated, it divides the 1.6 GHz±k*4 MHz signal to a 200 MHz±k*500 kHz clock signal, in which the nominal frequency of fCLK is 200 MHz. Note that if fCLK is increased too much, then the circuitry of TV tuner chip 120 may not operate properly. If it is decreased too much, then DSP 130 and MCU 132 may not have enough bandwidth to perform the necessary processing tasks. Given these considerations, the firmware in MCU 132 allows the operating frequency of fCLK to range between 190 MHz and 207 MHz in 1 MHz increments. In other embodiments, the range and granularity of fCLK may each be either increased or decreased.
Thus clock synthesizer 126 is able to alter its output frequency in suitably fine increments over a range to move fCLK to a frequency that reduces harmful coupling and spur generation. It also does so using a low-cost standard 24 MHz crystal.
The sources of undesired signal coupling will now be described with reference to
Aggressor circuit 360 transfers energy through magnetic flux created by internal inductances, such as bond wires to external terminals. Inductances in the magnetically coupled circuits convert the magnetic flux into unwanted electrical signals. As shown in
Second, aggressor circuit 360 couples energy into the LO signal of synthesizer with VCO 340. This interfering energy creates spurs when mixed with signals in the VCO, which distorts the local oscillator signal and thus distorts IF OUTPUT.
Third, aggressor circuit 360 couples energy into crystal oscillator 330, which it mixes with internal frequencies to generate spurs.
If no frequency or harmonic falls within the passband, i.e. the frequency and all harmonics fall outside the passband, then the algorithm checks, at step 410, whether a spur auk within a desired range of crystal oscillator 330. The desired range is a range that doesn't produce any significant harmful spurs. If the spur does not fall in the desired range, then the flow returns to step 408 for the selection of a new fCLK frequency. If the spur does fall in the desired range, then the flow continues to step 412.
Step 412 determines whether the spur falls within the desired range of synthesizer 340. If so, then at step 414 the algorithm selects the current frequency as the operating frequency, and then frequency planning algorithm 400 ends at step 428.
If not, then algorithm 400 determines whether secondary criteria are met by checking, at step 420, whether the spur falls within a secondary desired band of synthesizer 340. As used herein, the secondary desired band is first a “taboo” channel (frequency band that is not allocated to any channel), or second, a channel in which a blocker channel is not the strongest, such as channels excluding the current channel ±6 channels or ±12 channels. If the spur falls within the secondary desired band, then algorithm 400 logs the frequency at step 422. If the frequency is the last frequency then algorithm 400 chooses, at step 426, the first frequency from the log as the operating frequency, and then ends at step 428. If not, then flow returns to step 408.
By performing the frequency planning associated with flow chart 400, TV tuner chip 120 is able to significantly reduce direct coupling and harmful spur generation. The frequency planning firmware in TV tuner chip 120 also has a mechanism to prevent the harmful effects of sub-harmonics, as will now be described with reference to
As described above, digital processor 124 produces a primary sub-harmonic at 200 MHz/16, or 12.5 MHz, when fCLK is set to its nominal frequency of 200 MHz, and additional sub-harmonics at M·fCLK/16. When a relatively low frequency channel spans a frequency of 50 MHz, 62.5 MHz, 75 MHz, or 87.5 MHz, the nominal value of fCLK creates a sub-harmonic in-band for N=4, 5, 6, and 7.
However TV tuner chip 120 allows fCLK to vary within a range of 190 MHz to 207 MHz around its nominal frequency. Thus
However there exists a critical frequency 550, which is about 100 MHz for TV tuner chip 120, at which the separation of the sub-harmonics is great enough for all sub-harmonics to be out of band for at least one available value of fCLK. Near the high end of the tuning range, TV tuner chip 120 generates a first sub-harmonic 630 for fCLK=190 MHz that is separated from a second sub-harmonic 540 for fCLK=207 MHz by a significant amount.
Accordingly, when the CHANNEL SELECTION signal indicates a channel that ties above critical frequency 650, TV tuner chip 120 attempts to select a value of fCLK to move all sub-harmonics out-of-band. When the CHANNEL SELECTION signal indicates a channel that lies below critical frequency 550, TV tuner chip 120 selects a value of fCLK that produces a sub-harmonic that is in-band but at a frequency that is tolerable.
Moreover the qualitative effect of interfering signals across an NTSC channel is also non-uniform, a fact which has been understood for quite some time.
Ideally fCLK could be selected to place all sub-harmonics out of the desired band. However this Objective is not always possible as described with respect to
Moreover it is possible to predetermine desired fCLK frequencies for each channel and to store these frequencies for later retrieval based on the selected channel. However adding enough storage space to save all these values on-chip is costly for a low-cost multi-standard integrated circuit. For example, there are 158 channels in NTSC alone, and TV tuner chip 120 supports other broadcast standards including PAL and SECAM as well as several variants of NTSC.
TV tuner chip 120 solves this problem by dynamically selecting this least objectionable frequency by approximating the characteristics of the “W” curve and using its available processing power to dynamically determine the least objectionable fCLK frequency for a channel that has been selected.
in another embodiment, the W curve can be expanded to take into account image frequencies. Since sub-harmonic energy at image frequencies can be mixed into the passband and not cancelled, this undesirable energy can be significant as well. Also other appropriate rules can be used for transmission systems other than NTSC. For example for North American digital television (ATSC) systems, the weighting function ensures that any spurious energy does not land on the pilot frequency, which is about 10 dB more sensitive than other in-band frequencies.
Thus TV tuner chip 120 achieves high integration and low cost by performing dynamic frequency planning. TV tuner chip 120 selects a frequency fCLK at which to operate integrated digital circuitry based on three coupling mechanisms: direct clock signal coupling, spur generation and coupling, and sub-harmonic coupling. TV tuner chip 120 dynamically selects a least objectionable frequency based on the channel selected by the user. In other embodiments, TV tuner chip 120 may implement frequency planning for other coupling mechanisms or for only one or more of the ones described herein in various combinations.
The above-disclosed subject matter is to be considered illustrative, and not restrictive, and the appended claims are intended to cover all such modifications, enhancements, and other embodiments that fall within the true scope of the claims. Thus, to the maximum extent allowed by law, the scope of the present invention is to be determined by the broadest permissible interpretation of the following claims and their equivalents, and shall not be restricted or limited by the foregoing detailed description.
This application is a division of U.S. application Ser. No. 12/641,623, entitled “Radio Frequency (RF) Receiver with Dynamic Frequency Planning and Method Therefor,” invented by Terry Dickey, Gerald Champagne, and Brian Mirkin, filed Dec. 18, 2009 and assigned to the assignee hereof, which is herein incorporated by reference in its entirety. Related subject matter is found in a copending patent application, application Ser. No. 12/641,540, entitled “Radio Frequency GM Receiver with Frequency Planning and Method Therefor,” invented by Sherry X. Wu, Mustafa H. Koroglu, Alessandro Piovaccari, and Ramin K. Poorfard, filed Dec. 18, 2009 and assigned to the assignee hereof.
Number | Date | Country | |
---|---|---|---|
Parent | 12641623 | Dec 2009 | US |
Child | 13524909 | US |